Skip to content

Jeffrey Wilkes: SuperKamiokande

March 15, 2011

Wilkes gave a review of the fantastic amount of information that was extracted from many years of data taking by the SuperKamiokande detector in the course of the last 15 years. I took sparse notes during his talk, but his slides are already available on the conference site. Here let me just give a short overview.

The present instantiation of SuperKamiokande (see right for a sketch of the detector) is version 4. The original one dates back to 1996, and is the one which produced the ground-breaking observation of atmospheric neutrino oscillations. The photomultiplier coverage back then was of 40%.

Version 2 came in four years later, when many dead PMT were replaced, and during the refilling of water in the tank occurred the famous implosion accident. They spent some time rebuiliding the detector, transferring surviving PMT into alternate positions, and finally achieving a 40% coverage again. The improvements had significant effects in the energy threshold achievable for solar neutrinos, but the improvement was less significant for atmospheric neutrino physics.

Then came SKIII, which still had the  original electronics, but enhanced PMT coverage. In 2008 the detector underwent another upgrade, where all the vintage front-end electronics was replaced it with contemporary hardware.

Wilkes showed one of the landmark results of SK: the high-statistics results for the fluxes of atmospheric neutrino as a function of zenith angles, which shows clearly the oscillation effect.

He then showed the current results of a two-flavour oscillation analysis on the delta m^2 – sin^2 (2 theta) plane, overlaid with the recent MINOS result. The results of the combined analysis give a consistent picture. Notably, these disfavour sterile neutrino disappearance at 7 sigma significance.

For the full 3-flavour oscillation results, they consider matter effects and a “solar term” simultaneously. The latter is a possible enhancement of electron neutrinos in the sub-GeV region. The 2-flavour and full-flavour analysis results are consistent; he observed that there is no significant change in the best-fit sin^2 (2 theta_23).

An interesting result -one which I had not been aware of, at least- was then shown by the speaker. Tau neutrino events can be searched in SuperKamiokande data. These interactions give rise a complicated event topology, which makes the identification of the leading lepton from tau decay difficult. They use a neural network, and since they expect a negligible primary tau flux, the goal is to detect tau neutrinos in atmospheric data, to test the “no tau appearance” hypothesis. This results in an excess from tau neutrinos (see figure on the left), a signal of over 200 tau-neutrino events. SuperKamiokande data are inconsistent with the “no tau appearance” at 3.8 standard deviations.

Another analysis seeks evidence for CPT vilation phenomena in  atmospheric neutrino data. They test whether neutrino and antineutrino mixing parameters are indeed the same. This is confirmed by the existing data.

They search also for rare processes. For nucleon decays, the lifetime of protons in the decay into pion-antineutrino pairs is >3.9E32 years, while for the decay of neutron to antineutrino- pizero is >1.1 E33 years. These limits are starting to challenge the predictions of the SO(10)  model, as is clear from the figure below, which shows the experimental limits for several decay modes (in years, on the x axis and shown by red bars) compared with theoretical predictions in blue.

For solar neutrinos, the method of detection by SuperKamiokande involves the fit of the Cherenkov rings that arise when relativistic electrons materialize in the detector. The energy resolution is not different for SKIII from what it was in SKI. From 548 days of live time, and an energy threshold of 6.5 MeV, they measure the Boron-8 flux, and the day-night ratio of -0.056+-0.031+-0.013.

5 Comments leave one →
  1. March 15, 2011 3:38 pm

    “Notably, these disfavour sterile neutrino disappearance at 7 sigma significance.”

    This sentence is confusing, due to a combination of a double negative, and a lack of clarity about what is disappearing, due to the lack of context in the abbreviated account (which, don’t misunderstand me, is greatly appreciated).

    I think that this sentence means to say that the data disfavor indications of a failure to detect the reduction in the quantity of neutrinos relative to theoretical expections that would expected to be seen if a sterile neutrino existed, at 7 sigma significance.

    If this reading is correct, I take it to mean that the Superkamiokande data imply that a sterile neutrino, if there is one, must be so heavy that mass-energy conservation prevents it from appearing at any detectable frequency in the system that Superkamiokande is observing. For practical purposes that would probably mean that the evidence strongly supports the proposition that are three and only three flavors of neutrinos and anti-neutrinos, because neutrinos are so light that one would not expect even a very heavy sterile neutrino to be so heavy that mass-energy conservation would prevent even a quite heavy one from appearing in the Superkamiokande data set and avoid detection at a seven sigma level.

    But, one could also read this sentence to say that the data rule out the hypothesis that there is no sterile neutrino at 7 sigma significance, since the signal suggesting a sterile neutrino has not disappeared. I have to think that such a result would be the headline of the story and receive much more discussion in this post if it were true, however, so I assume that this is a misreading of the sentence.

    Of course, maybe both of those readings are wrong and this sentence means something else entirely.

    Clarification of this point would be appreciated.

  2. March 15, 2011 4:09 pm


    indeed the sentence is confusing -it confuses me as well. I think that slide 7 of the presentation means to say that the data of atmospheric zenith angle fluxes are not compatible with alternative mechanisms of disappearance of neutrinos. The disappearance of neutrinos into sterile ones is then disfavoured by the data at 7 sigma level.

    I would encourage Mauro to explain this further, he is more expert than me on this issue.


  3. Mauro Mezzetto permalink
    March 15, 2011 11:33 pm

    OK, I know I should have presented the conference time ago and I’m not contributing to the blog as I promised to Tommaso. I have to say I’m too committed to the organization of the workshop, an engagement I grossly underestimated, and the dramatic events in Japan, with consequences to the experiment I’m working, T2K, and to the conference, are not helping in finding spare time. I just realized I followed only 4 talks today, and I missed the SuperKamiokande talk. I’m discovering that the organizers are not allowed, de facto, to follow the conference.
    Anyway what is the meaning of these 7 sigmas?
    SuperKamiokande (SK) discovered in 1998 that muon neutrinos (numu) disappear. In a three neutrino scheme they could become electron neutrinos (nue) or tau neutrinos (nutau). In the same years Chooz experiment at reactors excluded the possibility of conversions to electron neutrinos. So the only viable option are conversions to tau neutrinos … or to sterile neutrinos if we accept that neutrinos are more than three.
    SK measured that this numu disappearance happens at a difference of masses squared of about 3 milli-ev, and we know that the numu mass cannot be higher than about 2.5 eV.
    We are talking of steriles in this range of masses. SK cannot say much about steriles of high masses.
    How can SK discriminate disappearance to nutau from disappearance to steriles? In charged current interactions active neutrinos create a lepton; the largest part of the atmospheric neutrinos don’t have enough energy to create a tau lepton (with a mass of 1.77 GeV) : no way to separate nutaus from steriles with charged current events (steriles neutrino are such because they cannot interact with weak interactions).
    The only way to separate the two is by studying neutral currents, where the neutrino scatters and part of its energy is released to the target nucleus generating hadrons (protons, neutrons, pions etc). Neutral currents are a category of events much more difficult to detect. Nutaus can interact with neutral currents (no problems of tau mass since the neutrino just scatters) and steriles can’t.
    Data analysis disfavors disappearance in steriles at 7 sigmas.
    What is not said in the slides is the fact that one could imagine that the most of the muon neutrinos oscillate into tau neutrinos, but a fraction of them oscillates to steriles (subdominant oscillations). In the previous analysis the limit on the fraction of nutau that oscillate to steriles was around 20%, at 90% confidence level.
    There are more subtleties in the sterile analysis of SK data, they will be addressed by T. Schwetz talk of tomorrow and C. Giunti talk of Thursday.

    • March 16, 2011 3:27 pm

      Many thanks. Your explanation helps a lot (and believe me, I understand how the press of events can get ahead of blogging – the horror!). The extra detail on subdominant oscillations is especially appreciated.

  4. Shantanu permalink
    March 16, 2011 12:43 am

    Thanks for the nice summary.
    Maybe you or someone could also post the questions asked after every talk.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: